Fix bug in html2lrf that affected external links without a fragment.

installer.nsi

@@ -15,7 +15,7 @@ Var MUI_TEMP
 
 !define PRODUCT_NAME "libprs500"
 !define XPUI_BRANDINGTEXT "${PRODUCT_NAME} created by Kovid Goyal"
-!define PRODUCT_VERSION "0.3.14"
+!define PRODUCT_VERSION "0.3.15"
 !define WEBSITE "https://libprs500.kovidgoyal.net"
 !define PY2EXE_DIR "C:\libprs500"
 !define LIBUSB_DIR "C:\libusb-prs500"

src/libprs500/lrf/html/convert_from.py

@@ -343,7 +343,8 @@ class HTMLConverter(object):
             prop.update(self.parse_style_properties(tag["style"]))    
         return prop
         
-    def parse_file(self):        
+    def parse_file(self):
+        previous = self.book.last_page()
         self.current_page = Page()
         self.current_block = TextBlock()
         self.current_para = Paragraph()
@@ -358,6 +359,22 @@ class HTMLConverter(object):
             self.current_block.append_to(self.current_page)
         if self.current_page and self.current_page.get_text().strip():
             self.book.append(self.current_page)
+            previous = self.current_page
+        
+        if not self.top.parent:
+            if not previous:
+                previous = self.current_page
+            found = False
+            for page in self.book.pages():
+                if page == previous:
+                    found = True
+                if found:
+                    self.top = page.contents[0]
+                    break
+            if not self.top.parent:
+                print self.top
+                raise ConversionError, 'Could not parse ' + self.file_name
+                    
         
             
     def get_text(self, tag):
@@ -384,7 +401,8 @@ class HTMLConverter(object):
             para, tag = link.para, link.tag
             if not path or os.path.basename(path) == self.file_name:
                 if fragment in self.targets.keys():
-                    tb = self.targets[fragment]
+                    tb = self.targets[fragment]                    
+                    sys.stdout.flush()
                     jb = JumpButton(tb)
                     self.book.append(jb)
                     cb = CharButton(jb, text=self.get_text(tag))
@@ -401,7 +419,7 @@ class HTMLConverter(object):
                         self.files[path] = HTMLConverter(self.book, path, \
                                      font_delta=self.font_delta, verbose=self.verbose)
                         HTMLConverter.processed_files[path] = self.files[path]
-                    except Exception, e:
+                    except Exception:
                         print >>sys.stderr, 'Unable to process', path
                         traceback.print_exc()
                         continue
@@ -566,6 +584,7 @@ class HTMLConverter(object):
                             self.current_page.append(self.current_block)
                             self.current_block = TextBlock()
                     else:
+                        print 'yay'
                         found, marked = False, False
                         for item in self.current_page.contents:
                             if item == previous:

src/libprs500/lrf/pylrs/pylrs.py

@@ -456,6 +456,27 @@ class Book(Delegator):
         self.applySettings(settings, testValid=True)
     
 
+    def pages(self):
+        '''Return list of Page objects in this book '''
+        ans = []
+        for item in self.delegates:
+            if isinstance(item, Main):
+                for candidate in item.contents:
+                    if isinstance(candidate, Page):
+                        ans.append(candidate)
+                break
+        return ans
+    
+    def last_page(self):
+        '''Return last Page in this book '''
+        for item in self.delegates:
+            if isinstance(item, Main):
+                temp = list(item.contents)
+                temp.reverse()
+                for candidate in temp:
+                    if isinstance(candidate, Page):
+                        return candidate
+    
     def getSettings(self):
         return ["sourceencoding"]
     

src/libprs500/lrf/read_text_stream.py

@@ -13,17 +13,18 @@ def descrambleBuf(buf, l, xorKey):
         l-=1
     return a.tostring()
 
-f = open(sys.argv[1], 'rb')
-f.seek(0x0a)
-xorkey = struct.unpack('<H', f.read(2))[0]
-
-f.seek(int(sys.argv[2], 16) + 0x10)
-flags = struct.unpack('<H', f.read(2))[0]
-f.read(2)
-l = struct.unpack('<I', f.read(4))[0]
-f.read(2)
-raw = f.read(l)
-key = (l % xorkey) + 0x0f
-descrambled = descrambleBuf(raw, l, key) if (flags & 0x200) else  raw
-stream = zlib.decompress(descrambled[4:]) if (flags & 0x100)  else  descrambled
-print stream
+if __name__ == '__main__':
+    f = open(sys.argv[1], 'rb')
+    f.seek(0x0a)
+    xorkey = struct.unpack('<H', f.read(2))[0]
+    
+    f.seek(int(sys.argv[2], 16) + 0x10)
+    flags = struct.unpack('<H', f.read(2))[0]
+    f.read(2)
+    l = struct.unpack('<I', f.read(4))[0]
+    f.read(2)
+    raw = f.read(l)
+    key = (l % xorkey) + 0x0f
+    descrambled = descrambleBuf(raw, l, key) if (flags & 0x200) else  raw
+    stream = zlib.decompress(descrambled[4:]) if (flags & 0x100)  else  descrambled
+    print stream